Wearable health devices have skyrocketed in popularity in recent years. They help regular everyday people monitor their fitness goals, heartrate, blood pressure, sleep, and even can give you an estimate of stress levels. These wearable health sensors come in the form of watches, hidden in clothes, or even in something as small as rings (Han et al., 2022). They are also heavily incorporated in our health system. These devices are used to help medical professionals make decisions as they get live physiological measurements (Strangman et al., 2018). There have been incredible progressions made in wearable health devices, but unfortunately brain measuring systems have fallen behind. Ambulatory EEG is our best wearable system, mostly used for seizure detection. There are only a handful of brain measuring systems that use infrared spectroscopy (Strangman et al., 2018). The good news is that this fact leaves plenty of room for people to invent new tools to fill in the gap.
Creating A Wearable Brain Monitor
Introduction
Design
Wearable brain measuring systems have to be small, flexible, have a long enough battery life, be durable, and be able to read brain waves and electrical signals in the most uninvasive way possible (Mullen et al., 2015). This is a big ask, as your skull does a brilliant job of obscuring the signals neurons release. Wearable devices have to be durable enough to withstand day-to-day activities, sweat, humidity, and just general bumps and bruises. Then on top of all that, there is the appearance. No one wants to wear a wonky looking thing on their head all day. The best type of system would be able to hide in plain sight, camouflage in its surroundings, while measuring the brain data. These considerations are why I am proposing a microelectrode mechanical system that is woven into clips in hair extensions.
Hair extensors can be applied to anyone's head, and assuming they find a good match, can be very difficult to spot. This would allow the system to gather its data all day, while remaining barely noticeable. Hair extensions can come in a variety of forms, but for this system clip ins are the best. They have a long strip at the top of them (the root) that spans from one side of the head to the other, there is continuous contact from one side to the other, and you can have multiple in at one time, allowing readings to be gathered from different points on the head. EEG sensors would be used, as they allow a lot of data to be gathered from a tiny sensor. The sensors would be placed along the band of the Clip in extension, allowing readings to be collected from different areas of the skull. You could also place the hair extensions on different areas of the head, allowing the EEG to pick up signals from a variety of areas.
Powersource
Ideally, only one powersource is needed for the entire head of electrodes. The powersource would be a circular battery. Reducing scratches and stopping the patient from being poked. The powersource would be stuck to the back of the patient's neck. If the patient has longer hair it could be hidden but on people with short hair it would be exposed. The battery would have a sleek, protective cover. Protecting it from sweat, bumps, and its environment. The battery would attach to a person's neck using a silicone skin safe adhesive, keeping it in place for all day use.
The battery would be recharged by a passive energy source, body heat. This stops the patient from having to recharge the battery and keeps a continuous inflow of data. In order to collect the body heat and convert it into energy the patient must also wear a thermoelectric generator (Kim et al., 2021). Due to the limitations of the voltage production of thermoelectric generators a voltage step up converter must be integrated. The thermoelectric generator could be attached to a person's neck or shoulder area, keeping it close to the battery and allowing the patient to move freely without worrying about ripping or tangling wires.
Data Collection and Interface
Data collected from the dry electrode EEGs would be transferred to a small computer integration circuit. This small chip would be connected to the EEG hair extensions directly via the same type of wire. The circuit would then hold the data recorded in its memory. The chip would have the same waterproof cover as the battery, to ensure it can withstand its 24 hour use. For the comfort of the patient, it would be attached to the opposite (or same if you choose) side of the neck using the adhesive.
Ensuring the software for the chip can hold the data is important. Your brain does a ridiculous amount of things during the day, so special integrated software would have to be used. The software also has to synchronize the incoming data, while providing timestamps (Sugden et al., 2023). The data collected in the chip is then transferred to the patient's phone via bluetooth. This allows the patient to upload the data throughout the day so they can see live results. The data is uploaded into an easy to use app, allowing regular day to day people to see the different wavelengths and activity of their brain. Their phones provide accessibility, and the app would be free (I am not charging more than I have too).
Limitations
As interesting as this concept is, and as helpful as it has the potential to be, I am asking a lot of both the powersource and the data integration system. Such small parts have to play a huge role and our technology just isn't there yet. The battery and thermoelectric generator are probably unable to produce the amount of energy needed to power multiple strips of the hair extensions. They might be able to generate power for 1, maybe 2 (if weβre really stretching it). Thermoelectric generators are a fairly new concept, so they still have a long way to go, especially in regards to the compactness.
Although the design prioritizes camouflaging in, the hair poses a big problem for the EEG dry electrodes. They need to be flush against the skull and hair gets in the way. In order to get the best readings possible, you would need to create very clear parts in the hair and place the hair extension base directly onto it. This would be an extremely time consuming task. On top of this your hair moves as you go about your day, and it's very possible that hair gets under the electrodes. The extensions themselves can also move around, as clip-in extensions are often used since they're so easy to install and remove. However, this means that they are also more likely to move or even fall out. Both would interfere with the quality of the EEG data.
Marketing
The ideal market for these would be either the medical field, specifically in the use of Alzheimer's and brain injury, or to those who are really into health and fitness. From a medical standpoint wearable EEG has significant benefits, it can improve diagnosis accuracy, provide live data to practitioners, and give in depth insights (Dias & Cunha, 2018). There are so many benefits to live EEG readings, and the best use would be in the medical field.
However, these could also be marketed towards the rich yoga moms who care (maybe a bit too much) about their health, or to other health and fitness people who want a more in depth understanding of their cognitive health and function, without sacrificing their appearance. The market would be more directed at women, as the target market for hair extensions tends to be toward them. But, that doesn't mean men couldn't use them, short hair extensions could be made. The downside of that would be the power source and computer chip would be visible. In a perfect world wearable EEGs become accessible to everyone who needs them, as they could help treat, teach, and develop new strategies about the human brain.
References
- Byrom, B., McCarthy, M., Schueler, P., & Muehlhausen, W. (2018). Brain Monitoring Devices in neuroscience clinical research: The potential of remote monitoring using sensors, wearables, and mobile devices. Clinical Pharmacology & Therapeutics, 104 (1), 59β71. https://doi.org/10.1002/cpt.1077
- Dias, D., & Paulo Silva Cunha, J. (2018). Wearable health devicesβvital sign monitoring, systems and technologies. Sensors, 18 (8). https://doi.org/10.3390/s18082414
- Han, S. A., Naqi, M., Kim, S., & Kim, J. H. (2022). All β Day wearable health monitoring system. EcoMat, 4 (4). https://doi.org/10.1002/eom2.12198
- Jawd, S., Chekhyor, N., & Sabea, A. (2021). Types of Solar Cell Batteries and their Energy Charging Methods. Thermal Engineering and Applications, 8 (2), 16β22. https://doi.org/10.37591/jotea
- Kim, J., Khan, S., Wu, P., Park, S., Park, H., Yu, C., & Kim, W. (2021). Self-charging wearables for continuous health monitoring. Nano Energy, 79, 105419. https://doi.org/10.1016/j.nanoen.2020.105419
- Mihajlovic, V., Patki, S., & Xu, J. (2017). Noninvasive wearable brain sensing. 2017 IEEE SENSORS, 18 (19). https://doi.org/10.1109/icsens.2017.8234430
- Mullen, T. R., Kothe, C. A., Chi, Y. M., Ojeda, A., Kerth, T., Makeig, S., Jung, T.-P., & Cauwenberghs, G. (2015). Real-time neuroimaging and cognitive monitoring using wearable dry EEG. IEEE Transactions on Biomedical Engineering, 62 (11), 2553β2567. https://doi.org/10.1109/tbme.2015.2481482
- Strangman, G. E., Ivkovic, V., & Zhang, Q. (2018). Wearable brain imaging with Multimodal Physiological Monitoring. Journal of Applied Physiology, 124 (3), 564β572. https://doi.org/10.1152/japplphysiol.00297.2017
- Sugden, R. J., Pham-Kim-Nghiem-Phu, V.-L. L., Campbell, I., Leon, A., & Diamandis, P. (2023). Remote collection of electrophysiological data with Brain Wearables: Opportunities and challenges. Bioelectronic Medicine, 9 (12). https://doi.org/10.1186/s42234-023-00114-5
Other Course Work
π Assignment 1: EEG Analysis
Power spectrum - Tori Anderson.png
Raw vs filtered data - Tori Anderson.png
π¨ Assignment 2: BrainImation
π― Midterm Project
π Anderson_Midterm 1 - Tori Anderson.pdf
View Original PDFTori Anderson
1776467
Midterm 1
Part 1
For part 1 I created a flower garden. Each flower represents a different brain wave type; Pink = Alpha,
Blue = Beta, Theta = Purple, Gamma = White, and Delta = Yellow. Each flower grows taller based on
how prominent their respective brain wave is. The user's goal is to get the pink flowers to be the tallest,
as that means your alpha waves are most prominent and youβre in a calm awareness state. This program
could be used as a form of meditation guide, as the user is attempting to make their alpha waves more
prominent and therefore getting into a calmer state. The sun remains covered by clouds until the pink
flowers are the tallest, then the sky clears to make a beautiful sunny day. Unfortunately, beta waves are
the most prominent in the simulated data, so I wasnβt able to show the clouds disappearing. I learned a lot
while implementing it. I started being able to recognize where I can add new chunks of code and remove
chunks I didnβt like without causing errors, which was a huge accomplishment for me. I also started to
recognize some of the patterns that show up in the code, so I was able to fix some errors without AI help.
Part 2
For part 2 I am measuring the N170 component of ERPs. I am comparing the N170 for upright faces
compared to the N170 for Inverted faces and then a basic square. I have highlighted the N170 section for
both the Face stimulus and the Inverted face stimulus, so it's easier to see the similarities and differences.
Each face stimulus is paired with an object so you have a difference to compare to, this should help show
how big the N170 spike is. Each stimulus is being randomly generated, so the participant is unable to
recognize a pattern, and you get a more honest response. The averaging is done with the last 20 trials,
which is the max amount you can store. The averaged brain waves are the same colours for both stimuli
so you can directly compare between the two stimuli. If real EEG was used I expect to see a larger spike
for the FACE stimulus then for INVERTED FACE stimulus or the object stimulus. This is due to the
N170 representing an early stage of face processing, so it makes sense that there would be a larger spike
for the upright face. Your brain needs to have a bit of a delay when the facial stimuli is upside down or
inverted so it makes sense that the N170 is slightly smaller. The object stimulus has the smallest N170,
since a square doesnβt have any facial features.
π₯ Midterm 1 Part 2 - Tori Anderson.mov
π‘ Videos require Google Drive access. Open in new tab if it doesn't load.
π₯ Neurostim - Midterm part 1 - Tori Anderson.mov
π‘ Videos require Google Drive access. Open in new tab if it doesn't load.
π Anderson_Midterm_part2 - Tori Anderson.txt
π‘ Code is embedded in this portfolio - opens instantly in the live BrainImation editor (no internet required!)
π Anderson_Midterm_part1 - Tori Anderson.txt
π‘ Code is embedded in this portfolio - opens instantly in the live BrainImation editor (no internet required!)